Multi-band speech recognition in noisy environments

Okawa, Shigeki; Bocchieri, Enrico; Potamianos, Alexandros

doi:10.1109/icassp.1998.675346

Cited by 96 publications

(57 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This makes pink noise much more difficult for the codec's noise removal algorithm to filter, and therefore should influence the choice of bit rates in the packets. Furthermore, the use of such additive noise generation techniques is common practice for exploring the impact of noise on speech recognition methods (e.g., Tibrewala and Hermansky [1997], Okawa et al [1998], Junqua et al [1994]). …”

Section: Robustness To Noisementioning

confidence: 99%

Uncovering Spoken Phrases in Encrypted Voice over IP Conversations

Wright

Ballard

Coull

et al. 2010

ACM Trans. Inf. Syst. Secur.

View full text Add to dashboard Cite

Although Voice over IP (VoIP) is rapidly being adopted, its security implications are not yet fully understood. Since VoIP calls may traverse untrusted networks, packets should be encrypted to ensure confidentiality. However, we show that it is possible to identify the phrases spoken within encrypted VoIP calls when the audio is encoded using variable bit rate codecs. To do so, we train a hidden Markov model using only knowledge of the phonetic pronunciations of words, such as those provided by a dictionary, and search packet sequences for instances of specified phrases. Our approach does not require examples of the speaker's voice, or even example recordings of the words that make up the target phrase. We evaluate our techniques on a standard speech recognition corpus containing over 2,000 phonetically rich phrases spoken by 630 distinct speakers from across the continental United States. Our results indicate that we can identify phrases within encrypted calls with an average accuracy of 50%, and with accuracy greater than 90% for some phrases. Clearly, such an attack calls into question the efficacy of current VoIP encryption standards. In addition, we examine the impact of various features of the underlying audio on our performance and discuss methods for mitigation.

show abstract

Section: Robustness To Noisementioning

confidence: 99%

Uncovering Spoken Phrases in Encrypted Voice over IP Conversations

Wright

Ballard

Coull

et al. 2010

ACM Trans. Inf. Syst. Secur.

View full text Add to dashboard Cite

show abstract

“…In both methods, a set of acoustic features are extracted separately from each band. The difference is that while in standard spectrotemporal processing the features are concatenated and classified together (hence this approach is sometimes referred to as 'feature recombination' [12]), in multiband processing these features go into different classifiers, the outputs of which are combined into one score by the recombination unit. The multi-band processing scheme provides a lot of options as regards the band-level classifier, the level of recombination, and the recombination method used.…”

Section: Introductionmentioning

confidence: 99%

Robust Multi-Band ASR Using Deep Neural Nets and Spectro-temporal Features

Kovács

Tóth

Grósz

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Spectro-temporal feature extraction and multi-band processing were both designed to make the speech recognizers more robust. Although they have been used for a long time now, very few attempts have been made to combine them. This is why here we integrate two spectrotemporal feature extraction methods into a multi-band framework. We assess the performance of our spectro-temporal feature sets both individually (as a baseline) and in combination with multi-band processing in phone recognition tasks on clean and noise contaminated versions of the TIMIT dataset. Our results show that multi-band processing clearly outperforms the baseline feature recombination method in every case tested. This improved performance can also be further enhanced by using the recently introduced technology of deep neural nets (DNNs).

show abstract

“…In this paper, we will use the term sub-band processing. The positive benefits of this new approach to speech recognition are starting to be investigated and reported (Bourlard and Dupont 1996;Hermansky, Tibrewala, and Pavel 1996;Tibrewala and Hermansky 1997;Hermansky and Sharma 1998;Okawa, Bocchieri, and Potamianos 1998;Morris, Hagen, and Bourlard 1999). There is every reason to expect that sub-band processing might also profitably be applied to speaker recognition, improving prospects for realworld applications.…”

Section: Introductionmentioning

confidence: 99%

Obituary: W A (Bill) Ainsworth

2003

Computer Speech & Language

View full text Add to dashboard Cite

A growing body of recent work documents the potential benefits of sub-band processing over wideband processing in automatic speech recognition and, less usually, speaker recognition. It is often found that the subband approach delivers performance improvements (especially in the presence of noise), but not always so. This raises the question of precisely when and how sub-band processing might be advantageous, which is difficult to answer because there is as yet only a rudimentary theoretical framework guiding this work. We describe a simple sub-band speaker recognition system designed to facilitate experimentation aimed at increasing understanding of the approach. This splits the time-domain speech signal into 16 sub-bands using a bank of second-order filters spaced on the psychophysical mel scale. Each sub-band has its own separate cepstral-based recognition system, the outputs of which are combined using the sum rule to produce a final decision. We find that sub-band processing leads to worthwhile reductions in both the verification and identification error rates relative to the wideband system, decreasing the identification error rate from 3.33% to 0.56% and equal error rate for verification by approximately 50% for clean speech. The hypothesis is advanced that, unlike the wideband system, sub-band processing effectively constrains the free parameters of the speaker models to be more uniformly deployed across frequency: as such, it offers a practical solution to the bias/variance dilemma of data modeling. Much remains to be done to explore fully the new paradigm of sub-band processing. Accordingly, several avenues for future work are identified. In particular, we aim to explore the hypothesis of a practical solution to the bias/variance dilemma in more depth.

show abstract

Multi-band speech recognition in noisy environments

Cited by 96 publications

References 10 publications

Uncovering Spoken Phrases in Encrypted Voice over IP Conversations

Uncovering Spoken Phrases in Encrypted Voice over IP Conversations

Robust Multi-Band ASR Using Deep Neural Nets and Spectro-temporal Features

Obituary: W A (Bill) Ainsworth

Contact Info

Product

Resources

About